YouTube video ranking by aspect-based sentiment analysis on user feedback

Abstract

In the current big data and Internet of things world, a huge amount of data is available in the form of multimedia, shared on Web 2.0 platforms. Today, in the online smart learning process, the most widely used online video repository is YouTube, which also provides a place for users to express their views about videos in the form comments and rating. These user comments are still unexplored in the context of ranking in retrieval. Majorly, rating based on like/dislike is the only criterion considered to decide the relevancy and quality and in determining the ranking of the videos. Sometimes, the relevancy of the video is unveiled after watching the video, and as a result of this, irrelevant videos may be ranked higher. In this paper, we have investigated the impact of different aspects of the video’s subject from the user comments in the video retrieval using aspect-based sentiment analysis.